Skip to main content

Full text of "The Philosophy of Glacier Motion"

See other formats


Early Journal Content on JSTOR, Free to Anyone in the World 

This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in 
the world by JSTOR. 

Known as the Early Journal Content, this set of works include research articles, news, letters, and other 
writings published in more than 200 of the oldest leading academic journals. The works date from the 
mid-seventeenth to the early twentieth centuries. 

We encourage people to read and share the Early Journal Content openly and to tell others that this 
resource exists. People may post this content online or redistribute in any way for non-commercial 

Read more about Early Journal Content at 
journal-content . 

JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people 
discover, use, and build upon a wide range of content through a powerful research and teaching 
platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit 
organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please 


The Philosophy of Glacier Motion [pp. 481-501] 

Please Note: Due to an error in the print volume, the page numbering 
in this article skips back and repeats page numbers 497-500. Although 
the printed page numbers are repeated, the content is different between 
the two sets of pages having the same page numbers. We have 
indexed the pages in correct reading order, with the article content 
presented in its entirety. 

Please click on "Next Page" (at the top of the screen) 
to begin viewing this article. 




Having described the development of snow into ice 
that takes place in the neve of a glacier, Prof. James 
Geikie goes on to say : " Thus solidified and apparently 
rigid, one would at first suppose that hardened snow or 
ice would be as immovable as the rock of the mountain 
upon which it reclined. We know that a bed of tough 
clay will rest upon a considerable slope without sliding 
downward, and even the loose stones and debris which 
cover so many hillsides in a highland country find repose 
upon an incline of 30°. At Fourneaux the debris shot 
from the mouth of the great Cenis tunnel forms a still 
steeper slope. Mr. Whymper tells us that its faces have 
as nearly as possible an angle of 45°. But ice, which is 
a much more rigid body than even the hardest clay, will 
move upon a slope that is inappreciable to the eye."* 

Prof. Croll, in criticising Tyndall's Regelation The- 
ory, also says in this connection : " I presume that few 
who have given much thought to the subject of glacier- 
motion have not had some slight misgivings in regard to 
the commonly received theory. There are some facts 
which I never could harmonize with this theory. For 
example, boulder clay is a far looser substance than ice ; 
its shearing force must be very much less than that of 

* " The Great Ice Age," pp. 32, 33. 

4S2 The Philosophy of Glacier Motion. 

ice, yet immense masses of boulder clay will lie immova- 
ble for ages on the slope of a hill so steep that one can 
hardly venture to climb it ; while a glacier will come 
crawling down a valley which by the eye we could hardly 
detect to be actually off the level."* It seems strange 
that such eminent glacialists as Geikie and Croll should 
compare "a bed of tough clay" or the loose stones and 
debris which cover so many hillsides, in their inert state, 
with the incessantly moving glacier. Both, to be sure, 
rest upon an inclined bed ; but here the analogy 
ceases. The boulder clay or the rocky debris repre- 
sent in the economy of nature lifeless, functionless or- 
ganisms — those that have already fulfilled their mission, 
or else at most are endowed with potential energy from 
past activity. How different with glaciers ! Stretching 
up into that region where the fall of snow during the 
year is largely in excess of that disposed of by evapora- 
tion or the occasional discharge of an avalanche, glaciers 
constitute one of the great links in the circulation of 
meteoric waters. The glacier then, unlike the debris or 
recumbent bed of clay, has its function — that of relieving 
the excess of snow and preventing an indefinite accumu- 
lation of the waters of the earth in their solid state upon 
the domes of the continents. Thus the glacier has its 
origin in, and is maintained by, this region of falling and 
fallen snow, its fountain head. Here, then, first of all, 
we should look for the cause of its motion. Geikie and 
Croll in the above statements evidently had regard only 
to the glacier proper, the "glacier d icou lenient" of Ren- 
du, ignoring the neve or "glacier riservoir." As well, 
however, enter into the discussion of the philosophy of 

* "Climate and Time," p. 497. 

The Philosophy of Glacier Motion. 483 

a river's motion and overlook entirely its source, as to 
inquire rationally into the motion of a glacier without 
due regard to its neve. Did the Mer de Glace, from its 
terminus at the sources of the Arveiron to its junction with 
its tributaries at Treleport, rest as at present upon its 
mountain bed, and all wasting away, from some conjunc- 
tion of physical causes, were at an end, then I contend 
even were it upon as steep a slope as the clay or rubble 
it would remain as immovable as they under like condi- 
tions. But of course, because of the physical properties 
of ice, these conditions could never be realized in nature. 
Let us then look at the matter from the opposite stand- 
point. Did the beds of clay and heaps of rubble exist 
under like conditions to the glacier, even upon far gentler 
slopes than those mentioned in the opening quotations, 
they would move and frequently have moved with fright- 
ful rapidity, carrying wide-spread devastation in their 
wake. The "mass of chalk on the Dorsetshire coast," 
that in 1839" slipped over a bed of clay into the sea," and 
the " thousands of tons of solid rock " that in the rainy 
summer of 1806, "suddenly swept across the valley of 
Goldau, burying four villages with about 500 of their 
inhabitants,"* will bear witness to this statement. Let, 
then, the heaps of " loose stones " receive constantly fresh 
accessions from the mountain-sides higher up, and the 
inertia of the mass will be overcome, its potential will be 
translated into an intense kinetic energy, and the entire 
mass, as it thunders down as an avalanche, will perform 
its function by relieving the mountain of its own wreck. 
Or let the clay have the cohesion of its constituent par- 
ticles overcome by saturation from percolating waters, 

* Encyclopaedia Britannica, Geology. 

484 The Philosophy of Glacier Motion. 

and the attachment of its bottom weakened by the under- 
mining and lubrication of underground waters, and the 
entire mass will be precipitated as a land slip. Both 
these conditions in a modified form would seem to exist 
in the glacier. The cohesion of its particles must become 
weakened by a partial melting ; its bed must become 
lubricated by the streams of water that come from the 
melting of its surface, while the falling flakes of snow, 
that by their aggregate weight first urged the glacier 
down the mountain-side, must help maintain its motion. 
Tyndall long since pointed out this analogy between 
moving earth and moving ice. In speaking of the curves 
that sweep across some glaciers on the union of trans- 
verse with marginal crevasses, he goes on to say : "In 
land slips, and in the motion of partially indurated mud, 
you may sometimes notice appearances similar to those 
exhibited by the ice."* The discussion of the physical 
cause of glacier-motion would seem upon analysis to have 
reached the following stage of development: 1. It has 
been proved again and again by most accurate experi- 
ments that, independently of its motion as a whole, the 
glacier experiences a differential motion — amotion of cer- 
tain of its constituent portions relatively to others : and 
further the same experiments have accurately defined 
just where the planes of swiftest motion lie, but it yet 
remains to be conclusively shown just what these portions 
are, their nature, size, and physical properties. 2. From 
all the known phenomena of glacier-motion there would 
seem to be two, and only two, of nature's great powers 
concerned in it — gravitation and heat. But it yet re- 
mains to be conclusively shown in just what way these 

* " The Forms of Water," pp. 108-9. 

The Philosophy of Glacier Motion. 485 

forces act. He who can give answers to these two ques- 
tions so correct that they will harmonize perfectly with 
each other and with all the known phenomena of glaciers 
and the laws defining their motion, will undoubtedly 
solve that problem which, up to the present time, has 
proved to be the most vexed in all physical science. It 
was the endeavor to do this satisfactorily that has led 
the most of our great geologists to formulate their now 
celebrated theories of glacier-motion. Let me, in a brief 
resume, notice the most prominent of these. 

After it became known that glaciers move — which 
was not until the eighteenth century — the most natural 
supposition of the early and superficial observer would 
be that they bodily slide over their beds as a slate would 
from a roof. Altman and Griiner, in 1 760, were the 
first to formally propound this "sliding" theory, which 
was afterwards revived, in 1 799, by the distinguished 
traveller and investigator, De Saussure.* In our own 
days Hopkins has so far modified this theory as to hold 
that it is the huge sections, into which the glacier seems 
to be divided by crevasses, that experience downward 

So early as 1773, Bordier of Geneva seemed to be 
struck by certain general resemblances between the mo- 
tion of a glacier and that of a river, and held that " the 
entire mass of ice is connected together, and presses 
from above downwards, after the manner of fluids," and 
that glacier-ice was like " softened wax, flexible and duc- 
tile to a certain point. "f In 184T, Rendu, the learned 
Bishop of Savoy, independently advanced similar views, 

* "Voyages," tome II. 

t " Picturesque Journey to the Glaciers of Savoy." 

486 The Philosophy of Glacier Motion. 

much more elaborate, however, as to detail. For he 
held that in glaciers, as in rivers, " the friction of the bot- 
tom and of the sides . . . causes the motion to vary, 
and " that " only towards the middle of the surface do 
we obtain the full motion." And further, that " glacier 
ice enjoys a kind of ductility which enables it to mould 
itself to its locality . . . as if it were a soft paste."* Un- 
fortunately, he had no exact measurements by which to 
verify his predictions. It was left for the celebrated 
glacialist Forbes, the following year, to apply a crucial 
test of Rendu's generalizations, and his own, with the 
aid of the theodolite. He then conceived the idea that 
" a glacier is an imperfect fluid, or viscous body, which 
is urged down slopes of certain inclination by the natural 
pressure of its parts ;" and that the molecules of ice must 
move over and past each other in their downward flow 
as those of water do in a river. Such is the substance of 
his well-known " viscous " theory. 

Recent investigations have shown an undoubted vis- 
cosity in ice, especially snow-ice. Helmholtz made an 
extended series of experiments showing that snow can be 
changed into ice by pressure, and that crushed ice can be 
moulded into almost any form. Herr Pfaff of Erlangen 
later made investigations in the same direction. Tyndall 
had attributed apparent viscosity in ice when under pres- 
sure to crushing and regelation. Bianconi of Bologna de- 
clared that while Tyndall's experiments showed that such 
might be the case where the changes of form took place 
rapidly, they did not preclude the possibility of ice -pos- 
sessing a small amount of viscosity. He conducted a 
series of experiments in 1871 on plates and bars of ice 

* " Theorie des Glaciers de la Savoie.". 

The Philosophy of Glacier Motion. 487 

submitted to bending and torsion. These experiments 
showed conclusively, that slow changes in form in ice 
can occur without any crushing and regelation, although 
the slightest jar during bending would shatter the ice- 
plates. Furthermore the lower or convex side of the 
bending plate, as Prof. Joseph Leconte has pointed out, 
can experience neither crushing nor regelation, as it 
is under tension, not pressure. Afterwards Messrs 
Mathews, Mosely, Tyndall, and Heim obtained similar re- 
sults. Prof. Bianconi and Heim further ascertained that 
granite pebbles and iron plates when slowly pressed into 
ice penetrated it as they would a viscous mass, the dis- 
placed particles of ice rising in a fringe about the intruding 
body. To preclude all possibility of regelation coming 
into play, Mr. Hungerford recently experimented with 
snow and ice under pressure, at a temperature ranging 
from 25° to 9° above O" F. and obtained results similar 
to Bianconi's. All this points to a certain plasticity 
in ice, whether this plasticity be concerned in the gla- 
cier's motion or not. 

Faraday long ago showed that two pieces of ice on 
being brought together would cohere or regelate at the 
points of contact. Tyndall calling attention to this 
property of ice holds in his well-known regelation theory, 
that it is the discrete particles of the glacier which ex- 
perience differential motion, having first been separated 
from the parent mass under strain, moved downwards 
on momentary relief of pressure, and finally regelated 
under renewed pressure by the constantly acting force 
of gravitation. Croll, struck by the fact that ice is dia- 
thermanous, concluded that the only way heat could 
pass through it would be by successive molecular melt- 

488 The Philosophy cf Glacier Motion. 

ings on the approach of the heat ray and resolidifications 
after its passage, the molecule of water giving out nearly 
the same amount of thermal energy that the molecule of 
ice had absorbed. In its liquid state the molecule occu- 
pying less space, and hence having room to move, would 
under the force of gravitation seek a lower level, there to 
resolidify and act as an entering wedge between adjacent 
molecules of ice. As glaciers must be penetrated by 
heat rays in all directions, Croll* concluded that the sum 
total of all such molecular transformations must consti- 
tute the distinct downward motion of the glacier. Such 
in brief is his Molecular theory. 

Carnot observed that pressure lowers the freezing 
point of water, and Prof. James Thompson found by ex- 
periment that this amounted to 0.0075° C. for every 
atmosphere of pressure. The latter heldf in his Pressure- 
liquefaction theory, based on this experiment, that the 
portions of a glacier at any instant, subjected to enormous 
but slowly applied pressure, would first liquefy, move 
downwards, and then resolidify, the pressure meanwhile 
being transferred to new portions. 

The earliest theory to be advanced, as far as is known, 
was by Scheuchzen of Zurich, in 1 705. He thought that 
the motion of a glacier must result from the expansion, 
at the moment of freezing, of the water in the body of 
the glacier, which percolated there through innumerable 
capillary fissures, from the melting surface above. J. de 
Charpentier of Bex, in 1841,$ again brought forward the 
same hypothesis, and so far elaborated it that it is now 

* " Climate and Time." 

f " Proceedings of the Royal Society," May, 1857. 

X " Essai sur les Glaciers," pp. 14, 103. 

The Philosophy of Glacier Motion. 489 

known as Charpentier's Dilation theory. L. Agassiz held 
the same views until he found* by experiment that the 
body of the glacier could not be the store-house of in- 
tense cold it was supposed to be, since, according to his re- 
sults, it has a mean temperature of 32°F., whether that of 
the surrounding air range far above or below this. M. 
Forel questions these results, however, as we shall see 
later. Canon Mosely observedf that sheets of lead, when 
placed upon an inclined plane of too gentle a gradient 
to be affected by gravitation alone, moved downward 
when subjected to changes of temperature. He attrib- 
uted this phenomenon to the superadded action of grav- 
itation tending to lower the centre of gravity of the 
sheets by always favoring the downward movement — of 
the upper end when they contract, of the lower end when 
they expand. This experiment was the basis of his modi- 
fied form of the dilation theory. 

Fr. Jos. Hugi, the distinguished naturalist of Soleure, 
was the first to describe the granules of glacier-ice,;}; 
which he designated by the name " cristaux du glacier" 
and the first to advance the Granular-dilation theory of 
glacier-motion. He recognized the fact that the granules 
increase in size from the neve downward, and to the ex- 
pansion force resulting from this growth he attributed 
the progress of the glacier. Ch. Grad later held similar 
views,§ and within the last few years Forel of Morges, 
has adopted and elaborated their generalizations in an 
article entitled " Le Grain du Glacier." j| The nature of 

* " Nouvelles Recherches," 1847. 

f " Proceedings of the Bristol Naturalists' Society" (1869) 

{ " Naturhistorische Alpenreise," p. 341. 

§ " Les Mondes," tome 35. 

J "Archives des Sciences," tome 7 (1882). 

490 The Philosophy of Glacier Motion. 

this granular growth, according to their hypotheses, is 
described by Forel in the same article : " It is the mol- 
ecular affinity which causes the crystal to increase in the 
mother water, in which it is plunged. The crystal is 
placed under such conditions that it must increase in 
volume ; it is bathed by the water at zero, which becomes 
colder ; this water cannot part with its heat without 
changing its state, when it passes into the state of ice ; 
this ice under the action of molecular forces, adds itself 
in new layers, in the same planes of crystallization, upon 
the periphery of the parent (Jancien) granule. The 
crystalline granule increases in volume."* Prof. Arnold 
Guyot, the great Swiss savant, also refers to the part the 
granules of the glacier play in its motion. " Glacier ice, 
however, never loses the traces of its origin, but a blow 
of the hammer will cause it to crumble to pieces and 
reveal its granular structure, "f 

Such are the principal theories that have been ad- 
vanced, stated as briefly as is consistent with accuracy, 
and without any reference to the warm and" often pro- 
tracted discussions that followed the presentation of 
some of them, notably Forbes's and Tyndall's. 

I will now assume with Hugi, Grad, and Guyot, that 
it is the component granules of the glacier that experi- 
ence differential motion, and then deduce from the laws 
and phenomena of glaciers the part, it seems to me, heat 
and gravitation must play in this motion. I have en- 
deavored to show in the opening paragraphs of this 
article that it is of vital importance in this discussion 
to consider most carefully the forces at play at the 

* "Archives des Sciences," tome 7, p. 351. 
f "' Physical Geography," p. 94. 

The Philosophy of Glacier Motion. 491 

source of the glacier. Let us then take as a typical 
glacier the Mer de Glace, and have regard to its principal 
source at the base of the Aiguille du Geant. The 
famous Col du Geant here constitutes the vast amphi- 
theatre where year after year is collected the snow that 
falls upon its surface, and that which is shed from the 
rocky pinnacle which gives it its name. As the at- 
mosphere of the Alps is moisture-laden, the snow-falls 
are in consequence large, and the waste by evaporation 
comparatively small, since the rays of the sun must often 
be screened off from the surface of the snow by clouds. 
The necessity then for some competent form of relief 
of the ever-accumulating mass of snow becomes patent, 
as the occasional discharges by avalanches are impo- 
tent to effect it. 

The architecture of these vast beds, of snow is clearly 
delineated at such points where a crevasse or berg- 
schrunde exposes to view a natural section. Here the 
snow has been observed by Tyndall and many other in- 
vestigators to be distinctly laminated, every pair of deli- 
cate blue bands traced along the snowy white mass 
defining the snow-fall for a given year. During the 
winter months the snow-fall is very large and of a dry, 
powdery nature. During the summer months it is small 
and of a moist, heavy nature, while the moisture is 
largely precipitated as fogs, mist and rain. In conse- 
quence, the snows of winter are gradually compacted 
and consolidated in summer by their superincumbent 
weight, aided by the various forms of precipitated waters 
and that which comes from the melting of the snow 

Prof. Tyndall has given the rationale of this process 

492 The Philosophy of Glacier Motion. 

of transformation of snow into ice. " At its origin, then, 
a glacier is snow — at its lower extremity it is ice. The 
blue blocks that arch the source of the Arveiron were 
once powdery snow upon the slopes of the Col du Ge- 
ant. Could our vision penetrate into the body of the gla- 
cier we should find that the change from white to blue es- 
sentially consists in the gradual expulsion of the air which 
was originally entangled in the meshes of the fallen snow. 
.... The snow which falls upon high mountain-emi- 
nences has often a temperature far below the freezing 
point of water. Such snow is dry, and if it always contin- 
ued so, the formation of a glacier from it would be impos- 
sible. The first action of the summer's sun is to raise the 
temperature of the superficial snow to 32°, and afterwards 
melt it. The water thus formed percolates through the 
colder mass underneath, and this I take to be the first 
active agency in expelling the air entangled in the snow. 
But as the liquid trickles over the surface of the granules 
colder than itself it is partially deposited in a solid form 
on these surfaces, thus augmenting the size of the gran- 
ules and cementing them together. When the mass 
thus formed is examined, the air within it is found as 
round bubbles. . . . The frost of the succeeding winter 
may, I think, or may not, according to circumstances, 
penetrate through the layer, and solidify the water which 
it still retains in its interstices. . . . 

" The ice of the neve at 32° may be squeezed or crushed 
with extreme facility ; and if the force be applied slowly 
and with caution, the yielding of the mass may be made 
to resemble the yielding of a plastic body. In the depths 
of the neve, when each portion of the ice is surrounded by 
a resistant mass, rude crushing is of course out of the 

The Philosophy of Glacier Motion. 493 

question. The layers underneath yield with extreme slow- 
ness to the pressure of the mass above them ; they are 
squeezed but not rudely fractured. . . . Thus, then, the 
lower portions of the neve are removed by pressure more 
and more from the condition of snow, the air bubbles 
which give to neve ice its whiteness are more and more 
expelled, and this process continued throughout the 
entire glacier finally brings the ice to that state of mag- 
nificent transparency which we find at the termination of 
the glacier Rosenlaui and elsewhere."* 

From data obtained by recent investigations con- 
cerning this evolution of snow into ice, M. Forel recog- 
nizes three distinct regions in a glacier, defining as many 
distinct stages of development : 

" 1. Ndvd (enfance du glacier). Excess of snow, the 
heat of summer not sufficient to melt the (entire) snow 
(fall) of the year. All the water produced is absorbed 
and assimilated by the profound icy layers : the temper- 
ature deep down (in the mass) much below 0°C. Ligne 
de separation. The heat of summer sufficient to melt 
all the snow of winter. But there is no excess of heat 
to attack the ice. 

" 2. Glacier adolescent. The heat of summer melts all 
the snow of winter and attacks by ablation a portion of the 
ice. All the infiltrated water is absorbed and assimilated 
by the ice : the temperature deep down in the mass much 
below 0°C. even at the end of summer. . . . The region 
of growth, the region of youth, in which the glacier is de- 
veloped. Ligne de separation. All the infiltrated water is 
absorbed by the increase of the granule of the glacier. 
Commencement of the glacial torrent at the end of sum- 

* "Glaciers of the Alps," pp. 249, 250, 251, 252. 

494 The Philosophy of Glacier Motion. 

mer. At the end of summer the temperature beneath 
the surface reaches 0°C. 

" 3. Glacier senile. The heat of summer is in excess ; 
the infiltrated water exceeds the quantity necessary for 
the reheating of the ice, which remains at 0°C, and the 
excess of water flows off into the glacial torrent The 
temperature of the ice remains at O o C. during summer. 
. . . The region of decrease, region of old age, when the 
glacier falls into decay."* The first line of separation 
is apparent upon the surface — the second is not. 

As the physiological forces at play in animal organ- 
isms can best be studied in their embryos, where they 
are least complicated, so the physical forces at work in 
a glacier can best be investigated in the embryonic 
neve, before the glacier d dcoulement has seen the light 
of day. Now no one, so far as I am aware, has ever wit- 
nessed the birth of a glacier, but from the data furnished 
by those questioned in their old age, one can arrive at 
some sort of idea as to what must take place. 

In the first place before the glacier, or even the par- 
ent neve, has appeared upon the scene, the climatic con- 
ditions of the region in which it is to have its birth must 
undergo a radical change. If previously dry, it must 
have its supernatant atmosphere laden with moisture. 
If moist before, it must at any rate have its mean annual 
temperature lowered to such a point that the moisture will 
be so largely precipitated as snow that the snow which 
falls will exceed that which is wasted. The once sunny 
skies are now frequently overcast with dense leaden 
clouds, the once genial air has taken on through a greater 
portion of the year an arctic temperature, while the oc- 

* " Le grain du glacier." Archiv. des Sc, tome 7 (1882) pp. 366-7. 

The Philosophy of Glacier Mo lion. 495 

casional downpour of a tropical shower has given place 
to long-continued and heavy falls of snow. 

Gradually the rains- and mists of summer, and the 
water from the melting snow, percolate through the dry 
mass. Gradually the now saturated beds become com- 
pacted in their lower parts by the pressure of those above, 
and consolidated by the frosts of winter. A part of the 
surface is licked up by the rays of the sun and the heated 
air, while from exposed rock surfaces huge shreds of snow 
are discharged as avalanches. But all this avails nothing 
in the face of the enormous falls of snow. 

" Supposing," says Tyndall, " two feet of snow a year 
to remain upon the Col, this would raise it to a height 
far surpassing that of Mont Blanc in five thousand years. 
Such accumulation must take place if the snow remain 
upon the Col. But the accumulation does not take 
place, hence the snow does not remain on the Col. The 
question then is, whither does it go ?"* Let us see. The 
accumulation evidently must go on until the cohesion of 
the mass is overcome by the enormous pressure brought 
to bear upon it. At this point and not before, the glacier 
first makes its appearance. When once started on its 
way down the mountain side, the glacier would descend 
below the snow-line to a point where its downward mo- 
tion is just counterbalanced by the melting of the ice. 
Should the mean annual temperature of the region be 
raised or its humidity lowered, from some conjunction of 
physical or cosmical causes, the glacier will shrink in di- 
mensions and the terminus will retreat up the mountain 
side. This retrogression has taken place during past 

' The Forms of Water," p. 49. 

496 The Philosophy of Glacier Motion. 

ages on an enormous scale, as old marginal moraines and 
stranded erratics clearly attest. 

Having now traced out the life-history of a glacier, if 
I may borrow this term from Biology, let us revert to 
the main point at issue — the investigation of the causes 
potent in the creation and maintenance of its motion. 
As the superficial portions of the neve are comparatively 
light and dry, it must be from the more profound layers 
that the icy tongue is thrust forth. The distinct strata, 
prolonged from the neve into the glacier, that can be 
traced in horizontal bands along the icy walls of some 
crevasses, furnish an absolute proof of this assumption. 
Whatever may be the case later in its development, it 
cannot be the weight of the glacier itself which drags it 
down, as none yet exists. We can account for its first 
appearance only by supposing that it is squeezed out from 
underneath the neve by the weight of snow and ice 
above it. 

When the glacier has attained its full size it seems 
natural, and I think that all the phenomena of its motion 
indicate, that the same forces that gave it birth must play 
a prominent part in its maintenance. We can, I think, see 
how the snow accumulating on the surface, and gradually 
compacting and settling down, must force out from its pro- 
founder depths fresh material to make good the terminal 
waste, and thus keep in motion the icy stream. This 
translation of downward pressure into a lateral thrust 
against the upper reaches of the glacier, causes the neve 
to act like a vast wedge. There is a limit to the glacier's 
power to resist its downward progress ; there would be 
almost none to the increase of pressure coming from an 
indefinite accumulation of snow upon the neve. The 

The Philosophy of Glacier Motion. 497 

force then which absolutely compels the glacier to move, 
nolens volens, whether upon a steep or a gentle grade, 
whether rigid from the intense cold of winter, or mobile 
from the heat of summer, is the incessantly acting and 
enormous pressure exerted by the snows accumulating 
upon the neve. If last winter's snow-fall has not pro- 
duced the necessary excess of pressure, next winter's will 
give another turn to the screw. In this vast ice-mill the 
power can be indefinitely accumulated — the resistance, so 
far from increasing, grows less. 

Pressure, as Prof. James Thompson has proved, tends 
to liquefy ice by lowering the freezing point of water. 
Pressure then must also produce a partial liquefaction of 
the granular neve, tending to weaken the cohesion of its 
particles and facilitate its motion. Thus gravitation at 
work in the neve is concerned in the motion of the 
glacier, directly by forcing it down the slope, and in- 
directly by increasing the mobility of its particles. 

Let us make a rough calculation what the total of 
this pressure must amount to in pounds. "It has been 
estimated," says Prof. Maury, "that the average annual 
snow-fall of the Alps amounts to sixty feet, which is 
equivalent to six feet of water."* Now a cubic foot of 
distilled water, at standard temperature and pressure, 
weighs a little over 74 pounds. So every year each 
square foot of the surface of an Alpine neve is put under 
afresh pressure of above 444 pounds : Or take a glacier 
with a neve surface of a quarter of a square mile, and 
the accession in pressure each year, from the snow-fall 
alone, would amount to the enormous sum of over 
1,547,251 tons. Let us go a little further: take 400 feet 

* " Physical Geography," p. 94. 

498 The Philosophy of Glacier Motion. 

as the average depth of the neve, and it would doubtless 
fall below the average of many ; then, as neve ice is 
"more than three times"* as dense as snow, the pres- 
sure exerted by the neve, upon its lower layers, and in 
consequence upon the upper reaches of the glacier, must 
be about 8,800 pounds, or over 4 tons to the square foot. 
Cut this estimate down one half, to allow for possible 
ever estimation of the density and depth of the neve, 
and we still have left a pressure of 2 tons to the square 
foot exerted against the upper cross section of the 
glacier. With such figures before us, I think it need no 
longer excite wonder that the neve, under the ceaselessly 
acting force of gravitation, can mould its own material, 
and become the principal factor in the glacier's motion. 

Heat, the other great engine at work, plays its part, 
by causing the saturation of the snow with the water 
that comes from the melting surface, and further must 
assist gravitation in weakening the cohesion of the neve. 
Both of these forces are at work in the glacier in a 
modified form to facilitate its downward motion. 

Before proceeding, however, let us first get a clear 
idea of the structure of glacier ice, and of the size, form 
and mode of aggregation of its component granules. 
" The glacier is therefore a mass of solid water of the 
special structure which mineralogists describe under the 
name of crystalline ; it is an agglomeration of crystalline 
granules locked one within the other, as are the granules 
of crystalline marble or of a lump of sugar. . . . The 
crystalline granule increases in size from the top to the 
bottom of the glacier's course : at the limit of the neve, 
the granule is larger than a small hazel nut ; in the 

* " Physical Geography," p. 95. 

The Philosophy of Glacier Motion. 499 

middle portion of a large glacier, it is the size of a wal- 
nut ; at the terminal portion that of a hen's egg. At 
the lower extremity of the Aletsch glaciers, the lower 
Aar and the Rhone, I have measured granules up to 7 
and 8 centimeters in their major diameter."* . . . These 
granules (cristaux) have not a regular form : they are 
irregular polyhedrals, locked (ench&sse's) one within an- 
other : the irregular curved faces of two neighboring 
granules are perfectly opposed to one another. So 
closely are these faces pressed the one against the other, 
that each granule retains the others in place and is itself 
retained by them, so that if " one succeed in disengaging 
one of the granules from a block of the glacier, then all 
the others disengage themselves more or less easily and 
the entire mass falls into separate pieces. "f 

So far we have treated of established facts. But 
when we would arrive at an accurate knowledge of the 
crystalline structure of the granules, we find that nothing 
as yet has been definitely determined Brewster was 
one of the first to make use of polarized light in the 
study of glacial ice. Then Tyndall, in his celebrated ex- 
periment, by condensing rays of heat in the interior of a 
block of ice, disclosed its beautiful six-sided crystalline 
structure ; for in the disk-like areas of fusion appeared 
the star-shaped "ice-flowers," as he called them. In 
1861, Sonklar came to the conclusion from his observa- 
tions that in a given granule the axes of crystallization 
were parallel, and hence that each granule must be a 
single crystal, but that between two neighboring crys- 
tals there was no uniformity of direction of the planes of 
crystallization. Bertin, on the contrary, in 1866 came 

* " Physical Geography," p. 95. f " Le grain rlu glacier." Arch, des Sci. tome 7. 

500 The Philosophy of Glacier Motion. 

to the conclusion that there is a certain determinate ar- 
rangement of the axes of crystallization, and that the ice 
of the glacier in its crystalline structure is little by little 
brought to resemble lake ice. Ch. Grad and A. Dupre 
advanced the same views in 1869. J. Miiller in 1872, 
and F. Klocke in 1881, differed from Bertin, and held 
that the planes of crystallization in glacier ice lay con- 
fusedly in all directions, which view Forel espoused in 
1882. Later in the same year Hagenbach-Bischoff pub- 
lished his views, in which he steers a middle course. On 
the one hand he differs from Klocke in that he believes 
that there is " a certain predominant orientation which 
is in the direction of pressure :" * on the other he would 
have Bertin, Grad and Dupre " replace the term parallel- 
isme de tous les axes by the more restrictive expression 
direction predominante." f 

Whatever may be true in regard to the planes of 
crystallization, it is certain that " the ice of the granule 
of the glacier is remarkably dense (compacte)," X while 
that which cements adjacent ones together must be loose 
in texture, principally from the bubbles of air located 
there. This is the reason why a blow of the hammer 
will cause "the ice" to crumble to pieces — break between 
the granules instead of across them. Now it is clear 
that whatever will bring about mobility among the 
granules will facilitate the motion of the glacier. And 
it is equally clear that the weakening of their cohesion, 
the great essential of mobility in all bodies, must come 
through a partial melting, which would of necessity 

* " Le grain du glacier," Archiv. des Sci., tome 8, p, 359. 

f Ibid., p. 360. 

% Ibid., tome 7, p. 332. 

The Philosophy of Glacier Motion. 497 

originate in the interstitial ice and remain there as long 
as it was partial. This melting of the interstitial ice, 
besides weakening the cohesion of the granules, must, by 
the very act, form about each a water-film which would 
of necessity facilitate the sliding of one past the other. 

Now gravitation, as we have already seen, produces 
this partial melting by lowering the freezing point of 
water under pressure. Pressure in the glacier must orig- 
inate in four ways : that which comes from the down- 
ward thrust of the ice masses higher up ; that which 
comes from the elevation of the glacier's head above the 
level of its lower reaches ; that which comes from its 
own superincumbent weight ; and finally, that which 
comes from the irregularities and friction of its bed. 

Heat, by causing the ice of the glacier to thaw and 
soften, also tends to produce mobility among its con- 
stituent particles by weakening their cohesion by a 
partial interstitial melting within the body of the glacier 
and by lubricating them with the water which has per- 
vaded the mass from the melting surface above. 

Having now given in detail the different ways in 
which heat and gravitation must act to produce motion 
in glaciers, let us see how this explanation of their com- 
bined action will harmonize with known glacial phe- 
nomena. As has been stated above the irregularities 
and friction of its bed must tend to produce pressure in 
a glacier and tension at right angles to this pressure, of 
which its veined structure and crevasses are an index. 
Thus a change of inclination of the bed of the channel, 
the widening or narrowing of its sides and the friction 
of both, produce respectively transverse, longitudinal and 
marginal crevasses and veins. Ice of the peculiar granu 

498 The Philosophy of Glacier Motion. 

lar structure to be found in glaciers, when under slowly 
applied but enormous pressure, aided by the heat of sun 
and air, will, like viscous bodies, move differentially, but 
on the slightest transition from the previously existing 
conditions of its channel, will, unlike these, experience 
fracture or compression as the case may be. The ice 
will pursue preexisting lines of motion, first as far beyond 
the point where the change takes place as the cohesion 
of the mass will resist tension, or pressure, and then 
sooner than yield will break or become compressed. 
In other words granular ice, owing to the physical con- 
ditions under which it exists in the glacier, experiences 
an almost perfect mobility of its constituent particles, 
within very narrow limits; but exceed these by a hairs- 
breadth, and it will lose its continuity or change its 
aggregate form sooner than yield to the stress put 
upon it. Tar and other viscous bodies, under like con- 
ditions, would experience a mobility of particles within 
very wide limits ; for in tar it is the ultimate molecules 
concerned in the motion — not the discrete granular par- 
ticles as in ice. The former is a true viscous body, the 
latter has been fitly called a viscoid body. 

In accounting for the erosion wrought by a glacier, 
it is only necessary to compare it with that of running 
water, remembering the difference between the free flow 
of water and the limited mobility and unyielding nature 
of glacier-ice. As the particles constituting the mass of 
the ice are mobile only within very narrow limits, we 
should expect it when in motion to wear down the in- 
equalities of its rocky bed, whether hard or soft, to one 
common level. On the other hand, the molecules of 
water, having an almost perfect freedom of motion in 

The Philosophy of Glacier Motion. 499 

seeking lines of least resistance, would spend most of 
their erosive energy upon the softer rocks, leaving the 
harder ones in bold relief. In like manner we should 
expect ice, when effecting erosion by means of rocks and 
earth held in suspension, to act, as Agassiz has expressed 
it, " like a vast file set in paste," creating ruts and 
scratches in the surface of its bed, parallel to the line of 
its swiftest motion, since the graving tools can of neces- 
sity experience no greater freedom of movement either 
upward or sidewise than the ice that holds them as in 
the grip of a vise. On the contrary, the foreign matter 
held in the slight and uncertain grasp of running water, 
curvetting hither and thither with every freak of its 
water-carrier, erodes irregularly and erratically. Fin- 
ally ice, from the strong cohesion of its mass, would be 
expected to pass over the small inequalities of its bed, 
like a moving bridge, only affecting those of larger area. 
Water would adapt itself to every irregularity of its bed, 
even the smallest. And so in every particular we find 
it in nature, as the polished billowy surfaces, parallel 
scorings and roches moutonne'es of old glacial regions and 
the irregularly carved rocks of old river beds, will re- 
spectively attest. 

That a glacier must move more slowly in winter than 
in summer becomes evident at once when we consider 
the part heat plays in its movement. Moreover, it is 
evident that the motion in summer must be exactly as 
much greater than that in winter, as the heat from all 
sources during the summer months exceeds that of the 
winter months, less the increase of motion that must of 
necessity result from the increased pressure of the win- 
ter's snow-fall. 

500 The Philosophy of Glacier Motion. 

That the friction of its bed should produce in viscoid 
ice the same differential motion it does in viscous tar is 
clear from the above reasoning. We should naturally 
expect that the granules near the centre and surface of 
the glacier, free from the friction and contact of another 
surface, would gradually gain on those retarded by the 
sides and bottom of the bed. It is also natural that the 
velocity of the glacier should increase with the slope, as 
increased slope means increased pressure and decreased 
resistance to its downward progress. 

That it should increase with the depth is also evi- 
dent, since that means increased pressure from the addi- 
tional weight of the superincumbent masses. 

That glaciers should conform to the larger and more 
gentle irregularities of the bed, and not to the smallest 
and sharpest, is clear from the above enunciation of the 
properties of granular ice. The cohesion of the usually 
rigid ice, under the stress of pressure and heat, is weak- 
ened far enough to allow it gradually to adapt itself to 
the larger inequalities of its bed, but under no circum- 
stances will it obtain mobility of particles sufficient to 
allow it, like fluid bodies, to suddenly accommodate itself 
to smaller inequalities. 

All the conditions discussed above become, of course, 
magnified in polar glaciers, as here we have huge sheets 
of ice that, as a general rule, disregarding the natural 
conformations of the ground, sweep over hill and dale 
alike, and finally push out into the sea, to give birth there 
to bergs and floes. These images of still vaster ice for- 
mations, in the past ages of our globe, must exist under 
an enormous pressure both from neve and glacier ; for 
the snow-falls on these lands of the midnight sun are 

The Philosophy of Glacier Motion. 501 

very heavy and of frequent occurrence through the 
greater part of the year, and, in consequence, the result- 
ing glaciers are of gigantic proportions. Some attain a 
thickness of several thousand feet, and, as in the case of 
the Humboldt glacier, a frontage of forty-five miles, 
while the Antarctic ice-cap constitutes the vast grave- 
stone of a continent forever dead to man. This enor- 
mous increase of pressure, as has already been stated, 
insures a higher rate of motion as compared with Alpine 
glaciers, amounting to as much as sixty feet per day. 

In conclusion, I would sum up by saying that it 
seems to me the action of gravitation, especially that at 
play in the neve, must be the most important factor con- 
cerned in glacier motion, while the action of heat, though 
essential for the evolution of snow into ice, must ever 
be regarded as subsidiary to the former. Gravitation 
acts uniformly all the time and under all conditions — 
heat can have but little influence in the dead of winter 
or in high latitudes ; yet it is just in these regions that 
glaciers experience the highest rate of motion. 

Occidental College, Los Angeles, Cal.